Page 1
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 195
GENETIC RELATIONSHIPS IN CUCURBITA MAXIMA DUCHESNE GERMPLASM
COLLECTED IN TUNISIA AS REVEALED BY MORPHOLOGICAL MARKERS
HAMDI Khawla, MOKRANI Khaoula, BEN-AMOR Jihen and TARCHOUN Neji
Laboratory of Vegetable crops/ High Agronomic Institute of Chott Mariem, Sousse, Tunisia 4042
ABSTRACT
Squash is one of the most important crops in Tunisia. It showed great diversity in morphological
characteristics, particularly leaf length and width, peduncle length, fruit weight, fruit shape,
floral scar ratio, skin thickness, flesh thickness, fruit colour and seed shape. In the present
research, 31 accessions of squash were collected from different provinces of Tunisia and
phenotypic diversity in their fruit characters was assessed. The collection showed appreciable
phenotypic variation in all the traits. Principal component analysis (PCA) and cluster were
performed to determine relationships among populations and to obtain information about those
fruit characters for the definition
of groups. The first two principal component axes accounted for 62.20% of the total multivariate
variation among the accessions. The greater part of variance was accounted for by floral scar
raio, leaf length, leaf width and flesh thickness. Cluster analysis identified four different groups.
The morphological characterization will be helpful to curators in the management and
improvement of squash germplasm in Tunisia
Keywords: Squash, morphological characterization, accessions, PCA, cluster
1. INTRODUCTION
Cucurbitaceae is a large family including many species with an economic interest such as melon,
watermelon, various squash and pumpkins. Many species are consumed in various forms, such as
seeds, leaves, fruits and sometimes flowers in villages all over Africa. The genus Cucurbita L.
(pumpkins and squash) is native to America where people cultivate it for more than 10000 years
(Smith, 1989).
Species of this genus have become, in the 16th century, familiar and important in many
European countries (Janick and Paris, 2006). Cucurbits have dispersed rapidly from Europe to
other continents (Decker-Walters and Walters 2000, Paris, 2016) and there are three
economically important species of Cucurbita, namely Cucurbita pepo L., Cucurbita maxima
Duch. and Cucurbita moschata Duch., which have different climatic adaptations and are widely
distributed around the world (Robinson and Decker-Walters, 1997, Wu et al., 2011). According
to FAOSTAT (2017), the world's most cucurbit-producing countries are China, India, Russia,
Ukraine, the United States of America and Iran with production of 7 297 540 respectively; 4 987
Page 2
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 196
123; 1 232 162; 1 104 550; 863 460 and 603 629 tonnes in 2014. Much of the production of
Cucurbitaceae was based on traditional local cultivars that have been maintained by farmers for
centuries, such as Spain, Latin America and the Caribbean Africa (Gwanama et al., 2000, Lira-
Saade, 2002, Ferriol et al., 2003).
For a long time, pumpkins have been used for traditional medicine in some countries such as
China, Argentina, India, Mexico, Brazil and Korea, since the pulpit of pumpkins and their seeds
are rich not only in protein but vitamins, antioxidants, carotenoids and tocopherols (Stephenson
et al., 2001).
In Egypt, summer squash is considered one of the most popular vegetable crops (Abd El-Hadi et
al., 2012). Pumpkin seeds are also used in culinary practices mainly in the southern regions of
Austria, Slovenia and Hungary (Murkovi'c et al., 1996). Turkey is also one of the main centers of
diversity for cucurbits grown because of their adaptation to various ecological conditions
resulting from natural selection and selection of farmers (Sari et al., 2008).
The genetic diversity of plant species must be evaluated primarily by the use of morphological
markers (through the descriptors). According to Smith and Smith (1989), the essential elements
in the classifications of cultivated species are morphological characters. These characters are
considered as a main descriptive tool in the characterization of morphotype collections. Indeed,
studies on phenotypic diversity within Cucurbita populations have been evaluated by Whitaker
and Robinson (1986) and Hernandez et al. (2005).
In this study we present an analysis of the genetic variability in C. maxima collected from the
major growing regions of Tunisia using the morphological descriptors.
2. MATERIALS AND METHODS
Plant material
Mature fruits and seeds of squash accessions that were usually propagated and maintained by
farmers themselves were collected. Moreover, thirty-one squash accessions were included in our
investigation, including 6 Kerkoubi type, 14 Batati type, 9 Bejaoui type and 2 Kalaaoui type
derived from many prospection’s missions during the harvest period from September 2014 to
Mai 2015 from seven localities belonging to the main cultivation regions in Tunisia. Three fruits
collected from the fields represent each accession. The collection and geographic position are
shown in figure 1, table 1 and 2.
Table 1: Season, type, number and origin of the local squash accessions collected
Collecte
season Type
Accessions
number Délégation Governorate Latitude (N) Longitude (E)
Altitude
(m)
Page 3
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 197
2014-
2015
Batati
Sahel
5 Chott-
Meriem Sousse
35°54'22.21"N
10°32’47.81"E
6
5 Sehline Monastir 35°45'05"N
10°42'39"E
10
Kerkoubi
2 Sidi-Bou-Ali Sousse
35°57'24"N
10°28'23"E
8
2 Sehline Monastir 35°45'05"N
10°42'39"E
10
2 Chott-
Meriem Sousse
35°54'22.21"N
10°32’47.81"E
6
Bejaoui
Sahel
3 Chott-
Meriem Sousse
35°54'22.21"N
10°32’47.81"E
6
3 Sehline Monastir 35°45'05"N
10°42'39"E
10
2015-
2016
Kalaaoui 1 Kalaat-
Andalous Ariana 37°033"N 10°11'7"E 2
Batati
Nord
2 Kalaat-
Andalous Ariana 37°033"N 10°11'7"E 2
3 Sidi-Hmeda Siliana 35°57‘28"N 9°32'57"E 800
Bejaoui
Nord
1 Kalaat-
Andalous Ariana 37°033"N 10°11'7"E 2
2 Sidi-Hmeda Siliana 35°57‘28"N 9°32'57"E 800
Total 31
Table 2: Local names and origins of the collected accessions
Page 4
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 198
Accession Local Name Origine
Ker1 Kerkoubi rouge Chott-Meriem
Ker2 Kerkoubi rose foncé Chott-Meriem
Ker3 Kerkoubi blanc Sidi-Bou-Ali
Ker4 Kerkoubi jaune Sidi-Bou-Ali
Ker5 Kerkoubi rouge Sehline
Ker6 Kerkoubi rose foncé Sehline
Bat1 Batati vert tacheté rose Chott-Meriem
Bat2 Batati jaune Chott-Meriem
Bat3 Batati blanc Chott-Meriem
Bat4 Batati orange clair Chott-Meriem
Bat5 Batati vert grisâtre Gazzeh (Sehline)
Bat6 Batati vert Gazzeh (Sehline)
Bat7 Batati blanc Bhira (Sehline)
Bat8 Batati orange clair Masjed-Aissa (Sehline)
Bat9 Batati jaune Masjed-Aissa (Sehline)
Bat10 Batati tacheté blanc Sidi-Hmeda (Siliana)
Bat11 Batati blanc Sidi-Hmeda (Siliana)
Bat12 Batati vert Kalaat andalous
Bat13 Batati vert grisâtre Kalaat andalous
Bat14 Batati orange clair Sidi-Hmeda (Siliana)
Bej1 Bejaoui vert Masjed-Aissa (Sehline)
Bej2 Bejaoui vert grisâtre Bhira (Sehline)
Bej3 Bejaoui blanc Bhira (Sehline)
Bej4 Bejaoui vert Chott-Meriem
Bej5 Bejaoui vert grisâtre Chott-Meriem
Page 5
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 199
Bej6 Bejaoui blanc Chott-Meriem
Bej7 Bejaoui vert Sidi-Hmeda (Siliana)
Bej8 Bejaoui vert claire Sidi-Hmeda (Siliana)
Bej9 Bejaoui kalâa Kalaat-Andalous
Kal 1 Kalaaoui Kalaat-Andalous
Kal 2 Kalaaoui Kalaat-Andalous
Morphological characterization
The collected fruits were brought back to the laboratory of vegetable crops of the Higher
Institute of Agronomic Sciences of Chott-Meriem, characterized based on IPGRI (International
Plant Genetic Resources Institute, 1998) and UPOV (International Union for the Protection of
New Varieties of Plants, 2007). The seeds were extracted, washed, dried, characterized and kept
cold. For the accessions represented by seeds, the latter have been characterized and preserved.
All seeds were treated with a fungicide and stored in cardboard packaging.
Twenty-two quantitative traits (cotyledon length and width, cotyledon ratio (L/W), stem
diameter, branches number, nodes number, leaf length and width, leaf ratio, male and female
flowers number, flower ratio, fruits number, fruit weight, peduncle length, floral scar length and
width, floral scar ratio (L/W), bark thickness, flesh thickness, seeds number, seeds length and
width) were measured with a ruler or caliper, fruit weight with a balance. Qualitative traits
(length of main stem, leaf blade size, intensity of green colour of skin, fruit shape in longitudinal
section, presence of neck, curving (longitudinal axis), profile at stem end, profile at blossom end,
grooves, distance between grooves, depth of grooves, marbling, main color of skin, intensity of
main color of skin, waxiness of skin, warts, Fruit: main color of flesh, seeds: color of coat) were
evaluated based on scoring and coding.
Statistical analysis
Data analyses were performed using the statistical procedures in SAS 9.1 software (SAS 1992).
Simple statistics such as means and coefficient of variation were used on quantitative parameters
in order to compare the variation between the studied accessions. A variance analysis (ANOVA)
was performed and then the averages were compared by Duncan’s multiple range test at 5%
level. A Pearson correlation analysis was carried out to estimate the relationship between the
studied variables. Different multivariate analyses were performed to evaluate the contribution of
each quantitative and qualitative character to the total variation: Principal component analysis
(PCA), factorial correspondence analysis (FCA) and hierarchical cluster analysis (HCA) were
conducted on quantitative, qualitative and mixed data respectively.
Page 6
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 200
3. RESULTS
3.1. Quantitative characterization
Variance of the morphological characters
Statistical analysis showed significant differences between accessions collected for all
parameters except skin thickness. Table 3 showed that leaf length and width vary respectively
from 37.50 to 55.33 mm and from 21.86 to 31.46 mm. Kerkoubi type accessions from Chott-
Meriem had the smallest leaves and the largest ones are found with the Bejaoui-Sahel type
accessions. The leaf shape has been described using the Length / Width ratio. It is found that the
Batati type of Sahel (Bat3, Bat4, Bat5, Bat6, Bat7, Bat8, Bat9 and Bat10) possess the higher
ratios which reaches 2.01. Their leaves are long rather than broad. The average value of the fruit
weight through the accessions is 7.60 kg and the average fruit weight of the individual accessions
varies from 4.55 to 10.90 kg. The Bejaoui type accessions (Bej2, Bej3, Bej4 and Bej5) have the
heaviest fruits. Concerning peduncle length, we noted that the Kerkoubi-Sahel type accessions
(Ker4, Ker5 and Ker6) had the shortest peduncles and the Bejaoui Sahel type accessions (Bej2,
Bej3, Bej4 and Bej5) had the longest peduncles. Moreover, the floral scar ratio varies between
0.43 and 0.80 with an average of 0.59. The lowest value was noted in the Kerkoubi type fruits,
so, we noted that it had a rounded shape. In addition, the highest value is found in the fruits of
Bejaoui, Batati and Kalaaoui types accessions whose fruits are long.
Concerning the skin and flesh thickness, we obtained high values (0.32 and 6.49 mm
respectively) in Kalaaoui-type originated form Kalaat-Andalous while Batati-Sahel type fruits
are characterized by lower values (0.11 and 3.92 mm respectively). In contrast, Batati type
accessions h the biggest seed producer (337 seed per fruit) and the smallest were Kalaaoui type
accessions (119 seed per fruit). According to the UPOV (2007) standards, seed ratios are all
lower or close to 0.5 which allows us to divide seeds between medium (Batati-Sahel and
Bejaoui-Sehline types accessions) and small (remaining accessions). Seeds length and width
varies from 1.26 to 1.97 cm and from 0.12 to 0.39 cm respectively.
Table 3. Means comparison for quantitative traits in 31 Tunisian squash accessions
Acc PL FW FLSL FLSW FLSR FTH STH SEN SEL SEW SER
Ker1 7.30bcd 1.13f 8.26c 34.07cde 70.40def 0.12c 0.49a 5.23b 188.67hij 18.20def 3.32cdefg
Ker2 6.86bcd 0.91f 3.62c 34.00cde 77.95def 0.26c 0.43a 5.81b 176.00hij 19.64bcd 3.36cdefg
Ker3 9.50abc 0.86f 7.29c 38.45cde 75.05def 0.13c 0.49a 4.45b 337.33a 12.72klm 3.87cde
Ker4 6.34bcd 1.66ef 9.67c 41.70cde 72.40def 0.31c 0.57a 3.92b 201.00ghi 12.62klm 3.88cde
Page 7
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 201
Ker5 10.90a 1.53f 11.33c 43.73cde 58.14def 0.22c 0.75a 4.81b 311.00ab 13.87klm 1.66efg
Ker6 7.13bcd 2.66def 6.07c 35.33cde 52.07ef 0.11c 0.69a 6.49b 119.00klm 14.26jkl 3.20efg
Bat1 5.13bcd 1.25f 8.17c 41.98cde 67.51def 0.15c 0.61a 5.43b 283.67abc 13.02klm 1.24fg
Bat2 6.27bcd 0.91f 9.00c 33.50cde 70.40def 0.33c 0.48a 7.79b 238.00efg 14.87ijk 2.69efg
Bat3 8.18bcd 1.20f 7.67c 33.54cde 63.03def 0.13c 0.53a 4.29b 276.67bcd 15.01hij 2.06efg
Bat4 9.20abc 1.40f 7.17c 49.14cde 82.51def 0.24c 0.59a 6.40b 262.33cde 18.41cde 3.76cde
Bat5 7.32bcd 1.08f 13.50c 43.15cde 57.39def 0.19c 0.75a 5.82b 73.33klm 20.22bcd 4.69b
Bat6 11.06a 0.33f 8.33c 37.10cde 70.47def 0.13c 0.52a 5.43b 102.00klm 16.81efg 2.31efg
Bat7 6.91bcd 1.66ef 6.83c 36.25cde 44.71f 0.13c 0.84a 5.71b 167.00ijk 16.97efg 3.23efg
Bat8 4.01cd 0.53f 7.00c 90.10abc 48.42f 0.43c 1.86a 6.71b 204.67fgh 20.35bcd 2.64efg
Bat9 7.41bcd 6.34ab 5.00c 44.86bcde 37.67f 0.13c 1.18a 5.38b 178.67hij 20.10bcd 3.75cde
Bat10 6.63bcd 0.46f 8.44c 64.33cde 46.57f 0.11c 1.38a 4.87b 174.00hij 20.00bcd 3.31efg
Bat11 8.15bcd 1.86ef 9.11c 89.98abc 52.41ef 0.34c 1.71a 2.83b 205.00fgh 20.59bc 3.46def
Bat12 6.83bcd 1.53f 5.84c 32.19de 46.69f 0.19c 0.69a 5.10b 117.00 klm 21.38b 4.17cd
Bat13 11.14a 2.80def 8.95c 50.06cde 44.04f 0.13c 1.13a 6.04b 177.00hij 21.51b 3.75cde
Bat14 8.59bcd 1.40f 7.66c 54.62cde 44.20f 0.21c 1.23a 6.58b 175.33hij 24.88a 11.60a
Bej1 10.66ab 1.60f 5.78c 13.68e 22.64f 0.11c 0.60a 2.86b 154.67jkl 15.64hij 3.50def
Bej2 6.80bcd 1.00f 10.22c 44.09cde 39.17f 0.29c 1.12a 5.87b 259.33def 11.81mn 2.47efg
Bej3 6.39bcd 5.93abc 75.52a 104.84a 151.21bcd 2.71ab 0.70a 43.39a 77.00klm 10.10n 2.49efg
Bej4 4.14bcd 4.54cd 58.95ab 96.11ab 148.79cde 4.02a 0.74a 28.93ab 67.00lm 16.30fgh 7.86b
Bej5 3.54cd 4.00cde 54.68ab 64.30cde 105.27def 2.83ab 0.44a 27.98ab 75.25klm 15.38ghi 7.22b
Page 8
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 202
Bej6 4.19bcd 6.18abc 31.84bc 64.01cde 105.27def 2.21abc 1.96a 35.73a 54.88m 13.87klm 1.59efg
Bej7 3.73cd 7.06ab 32.37bc 61.54cde 98.40def 1.59bc 1.37a 27.55ab 105.75klm 12.37lm 1.63efg
Bej8 6.97bcd 7.50a 53.16ab 86.60bcd 191.02ab 2.90a 0.51a 40.71a 123.63klm 25.44a 12.48a
Bej9 2.56d 5.92abc 69.57a 83.33bcd 191.03ab 1.89bc 053a 22.39ab 141.50klm 16.00ghi 11.20a
Kal 1 9.74abc 5.14bc 66.84a 104.14a 212.07a 2.85ab 0.54a 37.97a 129.00klm 16.73efg 3.30efg
Kal 2 8.59bcd 2.27def 59.73ab 75.26bcd 159.52abc 2.54ab 0.48a 42.52a 176.70hij 24.08a 11.98a
Note: ACC: accession, PL: peduncle length, FW: fruits weight, FLSL: floral scare length, FLSW: floral scare width,
FLSR: floral scare ratio, FTH: flesh thickness, STH: skin thickness, SEN: seeds number, SEL: seeds length, SEW:
seeds width, SER: seeds ratio
Principal components of morphological characters
Following principal component analysis, leaf length and width, peduncle length, and floral scar
ratio are the parameters with the highest contributions in the construction of the first axis with
values of 0.91, 0.89, 0.88 and 0.88 respectively. While the second axis is mainly composed of
the flesh thickness with 0.90% of contribution (Table 4).
Table 4. Principal component analysis (PCA) of characters associated with the squash (C.
maxima Duch.) accessions
F1 F2
Eigen value 8,771 5,534
Variabilité (%) 38,135 24,063
% cumulé 38,135 62,197
Traits Eigen vectors
LFL 0,917 0,007
LFW 0,892 0,168
LFR 0,441 -0,272
FW 0,471 -0,210
PL 0,883 -0,369
Page 9
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 203
FLSL 0,781 -0,370
FLSW -0,647 -0,663
FLSR 0,883 0,366
FTH 0,008 -0,499
STH -0,263 0,904
SEN 0,377 -0,597
SEL -0,631 0,261
SEW -0,641 -0,349
SER -0,323 -0,491
According to Figure 3, the first two axis have the highest eigenvalues (8.77 and 5.53 respectively
for F1 and F2). They reserve 62.20% of the total variability. The first axis occupies 38.13% of
the observed variability. It was positively correlated with leaf and width length and floral scar
ratio (l/L) and negative with seed length and width. It separates the accessions according to their
leaf and floral scars forms.
The second axis explains 24.06% of the total variability. It is negatively correlated with the skin
thickness, seeds number, and leaf ratio and positively correlated with flesh thickness and seeds
length. It permit to differentiate the accessions according to their flesh thickness.
Considering axis 1, two groups of accessions can be distinguished (Fig. 4). The first group
consists of accessions that had the longest broadest leaves and smallest seeds. Some accessions
in this group had a high floral scar ratio (Bej4, Bej5, Bej6, Bej8 and Bej9) and others have high-
weight fruits with long peduncles such as Bej1, Bej2, Bej3 and Bej7. The second group contains
accessions with small leaves and long seeds (Ker3, Ker4, Ker5, Ker6 and Bat1) or wide seeds
(Ker1, Ker2 and Bat2).
Following the second axis, the accessions were divided into two groups according to their flesh
thickness. A thick flesh, a thin skin and a few seeds production characterize the first group. It
contained six Batati type accessions originated from Siliana and Kalaat-Andalous (Bat11, Bat12,
Bat13, Bat14, Kal1 and Kal2). While the second group contained the accessions having thick
skin, accessions in this group were divided into two sub-groups; The accession Bat10 had a high
grain yield and the rest of accessions (Bat3, Bat4, Bat5, Bat6, Bat7, Bat8 and Bat9) had a flat
floral scar and broad leaves.
Page 10
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 204
3.2. Qualitative characterization
A Factorial Correspondence Analysis (FAC) was performed to detect associations and
oppositions between squash accessions and qualitative traits, by measuring their contribution to
total inertia for each factor. Table 5 shows the eigenvalue and the cumulative percentage of the
qualitative traits on the first three factors. Factor 1 accounted 25.80% of the total variance and a
positive correlation with fruit shape, presence of neck, neck shape and neck length, groove depth,
and waxiness of skin. Factor 2 accounted for 24.69% of the total variance and was positively
correlated with profile at stem end, profile at blossom end, main skin color, flesh color, and
warts.
Table 5. Definition of the first three factors of FCA on the base of morphological
qualitative characters of Tunisian squash landraces
F1 F2 F3
Eigenvalue 0,078 0,074 0,054
Percentage (%) 25,802 24,691 17,846
Cumulative percent 25,802 50,493 68,340
Character Eigenvalue
SH 0,447 -0,027 0,038
PNC 0,976 0,203 0,178
SNC 0,296 0,147 0,009
LNC 0,545 0,194 0,071
CV -0,077 0,116 -0,083
PSE -0,282 0,031 -0,122
PBE -0,073 0,174 -0,215
GV -0,052 0,117 -0,056
DGV -0,040 0,059 -0,046
DEGV 0,033 -0,036 -0,037
MR -0,284 -0,557 -0,087
Page 11
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 205
MCS -0,301 0,774 -0,394
IMCS -0,020 -0,097 -0,009
SCS -0,151 -0,771 -0,027
MCF -0,054 0,040 -0,130
WXS 0,232 -0,164 0,164
WR -0,576 0,336 1,082
CCS -0,209 0,126 -0,139
Note: SH: shape, PNC: presence of neck, SNC: shape of neck, LNC: length of neck, CV: curving, PSE: profile at
stem end, PBE: profile at blossom end, GV: fruit grooves, DGV: distance between grooves, DEGV: depth of
grooves, MR: marbling, MCS: main color of skin, IMCS: intensity of main color of skin, SCS: second color of skin,
MCF: main color of flesh, WXS: waxiness of skin, WR: warts, CCS: color of coat seeds
The scatter plot of the correspondence analysis in the plane defined by the first two factors
(Fig.5) divided the squash accessions on the basis of the qualitative characteristics into three
distinct groups. The first group (G1) included the accessions of site 2 (Sidi-Bou-Ali) and 3
(Chott-Meriem) characterized by fruits without necks, transverse broad elliptic shaped, slightly
depressed profile at stem end, grooves usually present (84.61%), marbling absent to very weak,
absence of waxiness and warts. In Sidi-Bou-Ali, 57.14% of skin fruits were light orange, 28.57%
cream and 14.28% yellow. While in Chott-Meriem, accessions had skin fruits colored with light
orange (33.33%), green (16.66%), yellow (16.66%), cream (16.66%) and orange brown
(16.66%). Seeds in this group are cream, yellow and brown with the percentages 38.46, 30.76
and 30.76 % respectively.
The second group contains accessions from site 4 (Sidi-Hmeda) and 1 (Sehline, Masjed-Aissa
and Behira) presented by fruits with large grooves. Accessions coming from Sidi-Hmeda had
skin fruit colored with 85.71% green, 9.52% orange brown and 4.76% brown. Seeds are 50%
cream colored and 50% brown colored. While, skin fruit in the accessions coming from Sehline
had primary color cream (50%), green (40%) and yellow (10%). Seeds are cream, green and
yellow (50, 40 and 10% respectively).
The third group includes accessions from site 3 (Kalaat-Andalus). Fruits in this group are pear-
shaped and had strong neck. Grooves were medium and skin fruit was green with cream seeds
(83.33%) or cream with brown seeds (16.66%).
Clustering of the Tunisian accessions
The coefficients of genetic similarity between the 31 accessions collected vary between 0.93 and
1. According to Figure 6, there are four groups of accessions with a similarity coefficient 0.93.
Page 12
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 206
Four accessions (Bat12, Bat13, Kal1 and Kal2) in the fourth cluster were well separated from the
rest of accessions. The first cluster C-I is subdivided into three sub-clusters. C-I-1 consists of
four Kerkoubi type accessions (Ker3, Ker4, Ker5 and Ker6) originating from Sidi-Bou-Ali and
Sehline. Long seeds and small leaves characterize these accessions. C-I-2 contained two
Kerkoubi-type accessions (Ker1 and Ker2) from Chott-Meriem and three Batati-type accessions
(Bat8, Bat9 and Bat10) from Sehline. A broad floral scar and a thin flesh characterize this sub-
cluster. C-I-3 consisted only of one accession (Bat7) from the Batati-type of Sehline, which had
a thick skin. The second cluster C-II is subdivided into three sub-groups. C-II-1 is composed of
four accessions (Bat2, Bat3, Bat4 and Bat5) of the Batati type originating from Chott-Meriem. It
had broad seeds, a thick skin and a thin flesh. The accessions (Bat6, Bej8 and Bej9) belonged to
the subgroup C-II-2. A thin flesh and a thick skin characterized this sub-group. C-II-3 contains
only the Batati-type accession Bat1 from Chott-Meriem. It has long seeds and small leaves.
The third cluster C-III is subdivided into four sub-groups. C-III-1 consists of four Bejaoui-type
accessions (Bej3, Bej4, Bej5, and Bej6) from Sehline with long leaves and broad floral scars.
Two Bejaoui-type accessions (Bej1 and Bej2) from Sehline belonged to the sub-group C-III-2.
High fruit weight and high grain yield characterize all of these accessions. Sub-cluster C-III-3
contains two accessions (Bat11 and Bat14) from the Batati type originating from Siliana. It had
thick flesh and low grain yield. C-III-4 contained only one accession (Bej7) from the Bejaoui
type originated of Siliana with long leaves, high weight and high grains yield.
The last cluster C-IV is subdivided into two sub-groups. It contained accessions with the thickest
flesh, the lowest fruit weight and grain yield. C-IV-1 contained three accessions (Bat12, Kal1
and Kal2) originated from Kalaat-Andalous. In addition, C-IV-2 is formed only of one Batati-
type accession (Bat13) from Siliana.
4. DISCUSSION
The morphological variability observed in this study is similar to several studies showing that, in
domesticated plant species, morphological differences are often based on agronomic traits
(Morimoto et al., 2005). Similarly, researchers have found that genetic diversity in local squash
populations is high, including the shape, size and color of fruit, the number and size of seeds and
flesh quality, color and thickness (Nerson et al., 2000, Ferriol et al., 2003, Paksoy and Aydin
2004). In another study by Liu et al. (2013), the morphological characterization of two
collections (Chinese and Russian) of squash revealed a great genetic diversity. In the
Cucurbitaceae family, the significant contribution of fruit and seed characteristics to
morphological variability has been reported for watermelon (Maggs-Kolling et al., 2000,
Gusmini, 2003), gourd (Morimoto et al., 2005), zucchini (Dey et al., 2006) and squash (Paris,
2001). Squash types with big seeds produce small fruits and fewest seeds number per fruit. Such
a negative correlation between fruit and seed size and seed number per fruit could be attributed
to regulatory behavior of plants, mainly because of limited external resources. The same
observations were reported for cucurbits by Nerson (2005) and Ban et al. (2006). For example,
Page 13
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 207
seeds length and width ranged between 1.26-1.97 and 0.12-0.39 cm. These results were near to
the dimensions found by Iré et al. (2003) who studied three species of cucurbitaceae (Citrullus
sp., Cucumeropsis mannii Naudin and Lagenaria siceraria). Seeds size reach 1.7 cm in length and
1 cm in width.
Principal Component Analysis (PCA) provided additional informations on the usefulness of
defined group traits. Knowledge of the variation found in a crop species and how it is distributed
is important for the development of breeding programs (Gil and Ron 1992, Balkaya and Ergün
2007).
With the greatest genetic diversity, squash have generally different characteristics, such as fruit
shape, fruit weight and leaf size (Mohanty and Mishra, 1999). Leaf shape and fruit weight are
used as an important characters on this work. In a similar work, Wehner (2008) used the leaf
shape as a discriminating criterion in watermelon. Nerson (2002) has shown that fruit weight is a
selection criterion between varieties.
In a similar study, PCA summarized the diversity of 20 morphological squash traits into three
major factors: leaf factor (represented by leaf size), fruit factor (represented by fruit width), and
quality factor (flesh thickness). These three main factors could be the priority indices in genetic
material screening and genetic selection (Du et al., 2011).
The genetic diversity of local varieties is part of the economic value of global biodiversity and it
has a primary importance for the future global productions (Stoilova et al., 2005). In addition, the
conservation and maintenance of these genetic resources constitute a source of biodiversity for
the breeding programs (Balkaya et al., 2005).
It was indicated that the useful strategies for ensuring efficient collection and management of
genetic material are estimation of the genetic diversity and the determination of the relationships
between collections (Bozokalfa et al., 2009).
The qualitative characterization proved that accessions have many significant differences. In a
similar study, of genetic diversity, Du et al. (2011) showed that C. moschata in China had a wide
variety of shapes, such as, round, oval and pear-shaped. It was probably due to the genetic
variability of the species to adapt to various agro-ecological conditions.
5. CONCLUSION
Cucurbita maxima Duchesne has a high importance in vegetable cultivation of Tunisia and it was
so far under-concerned in terms of collection and characterization. In this study, the collection
contributed to maximize the number of collected accessions from all productive zones in Tunisia.
During our research work, morphological characterization based on the quantitative and
phenotypic traits of the different collected accessions indicates relationships between accession,
geographical origin and morphological characteristics (proximity of ecotypes with the same
Page 14
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 208
geographical origin or morphological characteristics). This characterization also facilitate the
establishment of descriptive sheet of the characters of each accession.
Regarding the importance of some quantitative traits, we were able to distinguish accessions
representing interesting agronomic characters. Batati-yellow originating from Sehline is the most
seeds productive. The Bejaoui-type accessions from Sahel (Sehline and Chott-Meriem) had high-
weight fruits. In addition, Kalaaoui-type originating from Kalaat-Andalous and Batati-type
originating from Siliana are characterized by thick flesh and thin skin.
Qualitative characterization revealed that accessions coming from Sahel of Tunisia have
generally orange or crème skin fruits and yellow seeds. While accessions coming from Siliana
and Kalaat-Andalous had green skin fruits and cream seeds.
This large genetic diversity must be exploited in improvement programs affecting both yield and
quality. The generated information will be helpful to curators in the management and
improvement of Squash germplasm in Tunisia.
REFERENCES
Abdelhadi J, Wefstaedt P, Galindo-Zamora V, Anders A, Nolte I, et al., 2012. Load
redistribution in walking and trotting Beagles with induced forelimblameness. American Journal
of Veterinary Research. In press
Balkaya A., Kurtar ES., Yanmaz R., Ozbakir M., 2005. Investigation on collecting,
characterization and utilization of winter squash and pumpkin genetic resources in the Black Sea
region. (In Turkish). The Scientific and Technical Research (TUBITAK) Project No. 104 O 144.
Ankara, Turkey.
Balkaya A., Ergün A., 2007. Determination of superior pinto bean (Phaseolus vulgaris L. var.
Pinto) genotypes by selection under the ecological conditions of Samsun Province in Turkey.
Turk. J. Agric. Forest. 31(5): 335-347.
Ban, D., Goreta, S., Borosic, J., 2006. Plant spacing and cultivar affect melon growth and yield
components. Sci. Hortic. 109, 238–243.
Bozokalfa MK., Eiyok D., Turhan K., 2009. Patterns of phenotypic variation in a germplasm
collection of pepper (Capsicum annuum L.) from Turkey. Span. J. Agric. Res. 7(1): 83-95.
Decker-Walters DS. and Walters TW., 2000. Squash, p. 335-351. In: K.F. Kiple and K.C.
Ornelas (eds.). The Cambridge world history of food. Cambridge Univ. Press. U.K.
Dey, S.S., Singh, A.K., Chandel, D., Behera, T.K., 2006. Genetic diversity of bitter gourd
(Momordica charantia L.) genotypes revealed by RAPD markers and agronomic traits. Sci.
Hortic. 109, 21–28.
Page 15
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 209
Du X., Yongdong S., Xinzheng L., Junguo Z., Xiaomei L., 2011. Genetic divergence among
inbred lines in Cucurbita moschata from China. Scientia Horticulturae. 127: 207-213.
FAOSTAT, 2017, http://www.fao.org/faostat/
Ferriol M., Picó B., Nuez F., 2003. Genetic diversity of a germplasm collection of Cucurbita
pepo using SRAP and AFLP markers. Genetic Resources and Crop Evolution 50: 227–238.
Gil J., Ron AM., 1992. Variation in Phaseolus vulgaris in the Northwest of the Iberian Peninsula.
Plant Breed. 109: 313-319.
Gusmini G., 2003. Watermelon (Citrullus lanatus) Breeding Handbook. NC State University,
Raleigh.
Gwanama C., Labuschagne MT., Botha AM., 2000. Analysis of genetic variation Cucurbita
moschata by random amplified polymorphic DNA (RAPD) markers. Euphytica 113: 19–24.
Hernandez SM., Merrick CL., Eguilarte L., 2005. Maintenance of squash (Cucurbita spp.)
landrace diversity by farmer’s activities in Mexico. Genet. Resour. Crop Evol. 52: 697-707.
Irié A., Zoro Bi., Kévin K., 2003. Caractérisation botanique et agronomique de trois espèces de
cucurbites consommées en sauce en Afrique de l’Ouest : Citrullus sp., Cucumeropsis mannii
Naudin et Lagenaria siceraria (Molina) Standl. Biotechnol. Agron. Soc. Environ. 7 (3–4), 189-
199.
Janick J., Paris HS., 2006. The cucurbit images (1515–1518) of the Villa Farnesina, Rome.
Annals of Botany 97: 165–176.
Lira R. and Caballero J., 2002. Ethnobotany of the wild Mexican Cucurbitaceae. Economic
Botany. 56(4): 380–398.
Liu C., Ge Y., Wang D., Li X., Yang X., Cui C., Qu S., 2013. Morphological and molecular
diversity in a germplasm collection of seed pumpkin. Scientia Horticulturae. 154: 8–16.
Maggs-Ko¨lling GL., Madsen S., Christiansen JL., 2000. A phenetic analysis of morphological
variation in Citrullus lanatus in Namibia. Genet. Resour. Crop Evol. 47, 385–393.
Mohanty, B.K., Mishra, R.S., 1999. Variation and genetic parameters of yield and its
components in pumpkin. Indian Journal of Horticulture. 56 (4): 337–342.
Morimoto Y., Maundu P., Fujimaki H., Morishima H., 2005. Diversity of landraces of the white-
flowered gourd (Lagenaria siceraria) and its wild relatives in Kenya: fruit and seed morphology.
Genet. Resour. Crop Evol. 52, 737–747.
Page 16
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 210
Murkovic,M., Hillebrand,A., Winkler, W. J., Leitner, E., & Pfannhauser,W., 1996. Variabil-ity
of fatty acidcontent in pumpkin seeds (Cucurbita pepoL. ). European FoodResearchand
Technology,203, pp 216–219.
Nerson H. (2002) Relationships between plant density and fruit and seed production in
muskmelon. Journal of the American Society for Horticultural Science. 127 (5), 245-256.
Nerson NH, Paris HS, Paris EP., 2000. Fruit shape, size and seed yield in Cucurbita pepo. Proc.
Cucurbitaceae 2000. Eds. Katzirand N, Paris HS. Acta Hortic. pp. 227-230.
Nerson H., 2005. Effect of fruit shape and plant density on seed yield and quality of squash. Sci.
Hortic. 105: 293–304.
Paksoy M., Aydin C., 2004. Some physical properties of edible squash (Cucurbita pepo L.)
seeds. J. Food Eng. 65: 225-231.
Paris HS., 2001. Characterization of Cucurbita pepo collection at the newe ya’ar research center,
Israe¨l. Plant Genet. Resour. Newsl. 126: 41–45.
Paris H., 2016. Genetic Resources of Pumpkins and Squash, Cucurbita spp. Genetics and
Genomics of Cucurbitaceae. Part of the Plant Genetics and Genomics: Crops and Models book
series (PGG, volume 20), pp 111-154
Robinson RW. and Decker-Walters DS., 1997, Cucurbits. Crop Production Science in
Horticulture. Cab International, New York.
Sari N., Tan A., Yanmaz R., Yetiir H., Balkaya A., Solmaz I., Aykas L., 2008. General Status of
Cucurbit Genetic Resources in Turkey. Cucurbitaceae 2008. Proceedings of the IXth
EUCARPIA meeting on genetics and breeding of Cucurbitaceae (Pitrat M.ed.). INRA. Avignon,
France, pp. 21-32.
Smith HSC., Smith OS., 1989. The description and assessment of distance between inbred lines
of maize. II. The utility of morphological, biochemical and genetic descriptors and a scheme for
the testing of distinctiveness between inbred lines. Maydica. 34: 151-161.
Stephenson A.G., Hayes C.N., Johannsson M.H., Winsor J.A., 2001. The performance of
microgametophytes is affected by inbreeding depression and hybrid vigor in the sporophytic
generation. Sexual Plant Reproduction, v.14, p.77-83.
Stoilova T., Pereira MM., Tavares De Sousa MM., Carnide VI., 2005. Diversity in common bean
landraces (Phaseolus vulgaris L.) from Bulgaria and Portugal. J. Cent. Eur. Agric. 6(4): 443-
448.Wehner TC., 2008. Overview of the genes of watermelon, Cucurbitaceae, IX th EUCARPIA
meeting, May 21.04.2008, Avignon, France, 79-89.
Page 17
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 211
Whitaker TW., Robinson RW., 1986. Squash breeding. In: BASSETT, M.J. (Ed.). Breeding
Vegetable Crops. Westport, Connecticut: Avi, pp. 209-242, 584.
Wu J., Chang J., Wu Q., Zhan H., Xie S., 2011. Molecular diversity of Chinese Cucurbita
moschata germplasm collections detected by AFLP markers. Scientia Horticulturae. 128: 7–13.
Figure 1. Geographic distribution of Cucurbita maxima collection used in this study. The
names of provinces and locations are in bold.
Page 18
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 212
Figure 2. Fruits and seeds of different accessions collected from the different prospected zones
Note: (a) and (b) from Sidi-Hmeda Siliana, (c) and (d) from Kalaat-Andalous, (e) and (f) from Chott-Meriem, (g)
and (h) from Sidi-Bou-Ali, (i) and (j) from Sehline
Page 19
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 213
Figure 3. Projection of the quantitative characters on the plane generated by the first two axis
Figure 4. Diagram showing the relationships among the Tunisian accessions of C. maxima at the
basis the first two principal component axes using morphological characterization
Page 20
International Journal of Agriculture, Environment and Bioresearch
Vol. 4, No. 01; 2019
ISSN: 2456-8643
www.ijaeb.org Page 214
Figure 5. Scatter plot grouping of 31 Tunisian squash landraces based on the first two principal
factors of FCA
Figure 6. Dendrogram obtained from cluster analysis of thirty-one Tunisian squash landraces
using the UPGMA